[R-sig-eco] ANOSIM-Beginner

[Gavin Simpson]

On Wed, 2011-11-09 at 14:33 +0000, yy.loh at york.ac.uk wrote:
> Hi, 
> 
> I have a silly question about anosim in R, this is because i couldn't find 
> the answer in any R Help or yet to find one from google. i.e.: How to 
> interpret the result from summary(dune.ano) in anosim?
> 
> 
> > summary(dune.ano)
> 
> Call:
> anosim(dat = dune.dist, grouping = Management) 
> Dissimilarity: bray 
> 
> ANOSIM statistic R: 0.2579 
>       Significance: 0.016  <---May i know the null hypothesis for test?

The Null hypothesis is that there is no difference in, in this case, the
species composition of the dune meadows samples in the different
Management types.

> Based on  999  permutations
> 
> Empirical upper confidence limits of R:
>   90%   95% 97.5%   99% 
> 0.112 0.166 0.200 0.281 
> 
> Dissimilarity ranks between and within classes:
>         0% 25% 50% 75% 100% N <---The implications from this table. Between 
> 4 58.50 104.00 145.500 188.0 147 BF 5 15.25 25.50 41.250 57.0 3 HF 1 7.25 
> 46.25 68.125 89.5 10 NM 6 64.75 124.50 156.250 181.0 15 SF 3 32.75 53.50 
> 99.250 184.0 15

The table came out strangely. Here is what I get:

Dissimilarity ranks between and within classes:
        0%   25%    50%     75%  100%   N
Between  4 58.50 104.00 145.500 188.0 147
BF       5 15.25  25.50  41.250  57.0   3
HF       1  7.25  46.25  68.125  89.5  10
NM       6 64.75 124.50 156.250 181.0  15
SF       3 32.75  53.50  99.250 184.0  15

What this is showing is the ranks of over all dissimilarities for
quantiles of the dissimilarities in each "class". The Between row is for
dissimilarities for pairs of samples in different groups, whilst the
other rows are for the within group pairs.

Basically, the test assumes Between and within (the other rows) should
have similar distributions under the Null hypothesis. The function
assumes that the ranked within group dissimilarities have equal range
and median.

The table is presented so the user can evaluate whether the assumptions
are being met etc.

HTH

G

> 
> Thanking you guys in advance.
> 
> Sincerely Yours,
> J
> 
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
> 

-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%